Enhancing Human-Robot Collaborative Search Through Efficient Space Sharing with On-Demand Bidirectional Interaction

Abstract

Human-robot collaborative search tasks represent a growing field of research, focusing on scenarios where humans and robots coexist in a shared space with the objective of locating specific objects. The complexity of these tasks lies not only in the physical search itself but also in the efficient sharing of the search space between human and robot participants. This paper addresses the critical aspect of communication within human-robot teams, specifically the timing and content of inquiries regarding areas already searched or planned to be searched by each party. We propose innovative methods aimed at optimizing the decision-making process for robots on when to initiate dialogue and how to effectively manage inquiries from human counterparts, who may pose questions at unpredictable times. We develop a system that facilitates such an on-demand interaction for evaluating the proposed communication strategies, and conduct experiments within a virtual environment designed to mimic real-world conditions. In this simulated space, human participants interacted with the robot through avatars, enabling a controlled yet realistic examination of human-robot interaction dynamics. Our experimental results highlight the significant impact of effective dialogue on the efficiency of collaborative search tasks. We demonstrate that timely and relevant exchanges between humans and robots can substantially reduce search times and improve overall task performance. This study not only advances our understanding of human-robot collaboration but also lays the groundwork for future developments in communication protocols for cooperative endeavors involving humans and robotic systems.